THe Spotify dataset by year contains data from 1922 to 2021, collected by Yamac Eren using Spotify’s web API. The dataset contains information on acousticness, danceability, duration, etc.
When using this data set there are a few things which are both interesting and beneficial to note. First, the number of rows in the data frame is 100 which each correlate to the years from 1922-2021. Throughout the data frame, the average length of songs this century is 2.2729675^{5} ms. The most common key used throughout the century was the key of C. Interestingly, the year in which music was the loudest was 1922. Finally, the highest amount of danceibility recorded in the dataset was 0.6929043.
Below shows a table on the progression of music throughout the decades. It displays the averages of various categories for music in its respective decade. For example, in the 1970s, 40% of all music was acoustic.This table shows us that over the decades, music has gotten less acoustic, instrumental, and live. But, it has increased in danceability, energy, loudness, and tempo.
| Decade | Acousticness | Danceability | Energy | Instrumentalness | Liveness | Loudness (dB) | Tempo (BPM) | Key | Duration (minutes) |
|---|---|---|---|---|---|---|---|---|---|
| 1920 | 86.9% | 55.9% | 25.3% | 39.8% | 21.8% | -16.2 | 111 | F#/Gb | 3.2 |
| 1930 | 87.4% | 54.2% | 28.4% | 27.3% | 22.6% | -14.2 | 112 | E | 3.5 |
| 1940 | 87.8% | 47.3% | 25.7% | 37.8% | 22.2% | -15.2 | 107 | F | 3.7 |
| 1950 | 84% | 47.7% | 28.7% | 24.8% | 20.9% | -14.7 | 111 | D | 3.7 |
| 1960 | 62.6% | 49.5% | 41.3% | 15.8% | 20.9% | -12.7 | 115 | C | 3.5 |
| 1970 | 40% | 52.5% | 53.4% | 11.6% | 21.6% | -11.4 | 120 | D | 4.2 |
| 1980 | 29.9% | 54.6% | 59.5% | 12.2% | 20.5% | -11.2 | 121 | G | 4.2 |
| 1990 | 30.8% | 56.6% | 58.6% | 11% | 19.7% | -10.0 | 119 | F#/Gb | 4.1 |
| 2000 | 27% | 57.4% | 65.2% | 8.4% | 19.6% | -7.5 | 121 | F | 4.0 |
| 2010 | 26.4% | 59.7% | 62.9% | 8.8% | 18.9% | -7.5 | 121 | D | 3.8 |
| 2020 | 22% | 69.3% | 63.1% | 1.6% | 17.9% | -6.6 | 124 | C#/Db | 3.2 |
Below you will find different visualizations that show how music has played a role in U.S. history.
Below you will find a chart that shows a scatter plot visualization of how music popularity has changed throughout the years.
## `geom_smooth()` using formula 'y ~ x'
We chose to use a scatter plot to show data because it best demonstrated the trend in music popularity. We found that the overall music popularity has seen a continuous increase over the years and has continued to become a more prominent part of our culture. The scatter plot shows a strong positive correlation between the popularity of music (y-axis) over the years (x-axis) and we included a line of best fit to show the general trend.
We chose to use a stacked bar chart as the intent was to compare the overall change in “danceability” and “instrumentalness” of music over time, to see if there exist a relationship between “danceability” and “instrumentalness”. Based on the chart, we found that from 1921 to 2020, there seems to be no significant change in “danceability”. However, there is clearly a downward trend in “instrumentalness”.
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
We chose to use a line graph with a smooth line to clearly visualize the change in the loudness of music (dB) over time. The smoothline further aids the audience to capture the overall trend of the loudness level of music from 1921 to 2020. From our observation, we discovered that the change in loudness level from the early 1920s to 1950s was very dynamic, ranging from -19.275dB to -13.864. Also, the smooth line conveys that the overall loudness level of music has been gradually increasing, reaching its maximum level of -6.595067dB in year 2020.